Novel 1-phenylpiperazine and 4-phenylpiperidine ... - ACS Publications

ACS Legacy Archive. Cite this:J. Med. Chem. 34, 12, 3360-3365. Note: In lieu of an abstract, this is the article's first page. Click to increase image...
0 downloads 0 Views 858KB Size
J . Med. Chem. 1991,34,3360-3365

3360

Novel 1-Phenylpiperazine and 4-Phenylpiperidine Derivatives as High-Affinity 0 Ligands Richard A. Glennon,**tMamoun Y. Yousif,+Abd M. Ismaiel,+ Mahmoud B. El-Ashmawy,t J. L. Herndon,t James B. Fischer,t Alfred C. Server,* and Kathleen J. Burke Howie* Department of Medicinal Chemistry, School of Pharmacy, Medical College of Virginia, Virginia Commonwealth University, Richmond, Virginia 23298-0540, and Cambridge NeuroScience Research, Cambridge, Massachusetts 02139. Received April 18, 1991 u receptors may represent an exciting new approach for the development of novel psychotherapeutic agents. Unfortunately, many of the commonly used u ligands lack selectivity (e.g., many bind at phencyclidine or dopamine receptors) or suffer from other serious drawbacks. Recently, we described a series of 2-phenylaminoethanes that bind at u receptors with high affiiity and selectivity. Because there is evidence that 1-phenylpiperazinescan structurally mimic the 2-phenylaminoethane moiety, we prepared a series of 1-phenylpiperazines and related analogues and incorporated structural features already shown to enhance the u binding of the 2-phenylaminoethanes. Several of these derivatives bind at u receptors with high affiity (Ki= 1-10 nM)and lack appreciable affinity for phencyclidine and dopamine receptors. In as much as certain of these agents structurally resemble the high-affinity, but nonselective, u ligand haloperidol, and because they bind with 10 times the affinity of haloperidol, we have apparently identified what appears to be the primary u pharmacophore of that agent.

u receptors represent an exciting new area of research new group of u ligands, diaminocyclohexane derivatives, that is attracting attention for their possible involvement has been shown to bind with much greater selectivity.20 in various psychiatric disorders.1*2 Certain classical neuroleptics, such as the butyrophenones and the phenot h i a z i n e ~and , ~ ~ a~ variety of nonclassical n e u r ~ l e p t i c s ~ ~ ~ have been demonstrated to bind at u receptors. Members of several classes of antidepressants, such as the tricyclic antidepressants,6 monoamine oxidase inhibitors,6J and 1 2 serotonin uptake inhibitors: also bind at u sites. At this time, however, it is unclear whether these agents produce any of their therapeutic effects via a u mechanism. u ligands may also have indirect effects; for example, they may modulate N-methyl-D-aspartate functionDand glutamate release,1° perhaps via interaction at polyamine rec3 ognition sites," and dopaminergic function.12 u receptors have also been implicated in certain physiological functions such as colonic motility,13intestinal ion transport,14 synthesis of the pineal hormone melatonin,15and possibly in motor activity16and dystonic di~orders.'~Much of the present uncertainty regarding a precise role for u receptors stems from a lack of high-affinity a-selective ligands. Several different types of agents have been demonstrated to bind at putative u receptors, and the most commonly employed include a-opiates such as N-allylnormetazocine (NANM, SKF 10,047) and cyclazocine (1, R = CH2CH=CH2 and CH2-c-C3H5,respectively), the F 5 guanidine derivative 1,3-di-o-tolylguanidine(DTG 2), the 3-phenylpiperidine (+)-3-(3-hydroxyphenyl)-N-(1propy1)piperidine [ (+)-3-PPP; 31, and certain butyrophenone neuroleptics such as haloperidol (4). Unfortunately, many of these agents suffer from serious drawbacks NH-R that limit their utility as tools to investigate u pharma6 35 cology. The u opiates 1 possess only modest affinity for u sites and, in addition, bind with high affinity at phen(1) Junien, J. L.; Leonard, B. E. Drugs acting on u and phencycyclidine (PCP) sites.lv3 DTG (2) has been shown to clidine receptors: A review of their nature function and posproduce PCP-like stimulus effects in rats.l8 3-PPP (3) sible therapeutic importance. Clin. Neurophurmacol. 1989,12, and many of its analogues possess considerable dopami353-374. nergic character.lg Although haloperidol is reported to (2) Seeman, P. Atypical neuroleptics: role of multiple receptors, be one of the highest affinity u agents, it lacks selectivity endogenous dopamine, and receptor linkage. Acta Psychiat. Scand. 1990,82 (Suppl. 358), 14-20. and binds at D2 dopamine receptors with an affinity (3) Largent, B. L.; Wikstrom, H.; Gundlach, A. L.; Snyder, S. H. comparable to its affinity for u sites.2 Other butyroStructural determinants of u receptor affinity. Mol. Pharmaphenone neuroleptics, such as the arylpiperazine azaperone C O ~ .1988,32, 772-784. (5): also bind at u sites but these agents typically bind with (4) Snyder, S . H.; Largent, B. L. Receptor mechanisms in antipeven higher affinity at dopamine receptors. Recently, a sychotic drug action: Focus on sigma receptors. J. NeuropVirginia Commonwealth University. Neuroscience Research.

1 Cambridge

sychiat. 1989, 1 , 7-15. (5) Largent, B. L.; Wikstrom, H.; Snowman, A. M.; Snyder, S. H. Novel antipsychoticdrugs share high affinity for u receptors. Eur. J. Pharmacol. 1988,155, 345-347.

0022-262319111834-3360$02.50/0 0 1991 American Chemical Society

Novel 1-Phenylpiperazine and 4-Phenylpiperidine Derivatives

Journal of Medicinal Chemistry, 1991, Vol. 34, No. 12 3361

Table I. u Receptor Affinities of Some Simple Piperazine Derivatives

CH2CH2CH2C6H6) binds at u sites with high affinity (Ki = 20 nM and about 20 times the affinity of the a-opiate

A r - X 7 4 - R

W

X Ar R N H phenyl H benzoyl H 3-C1-phenyl H 3-CF3-phenyl H 11 4-C1-phenyl H 12 2-0Me-phenyl H 13 4-OMe-phenyl H 14 1-naphthyl H 15 2-naphthyl H 16 2-pyrimidinyl H N 17 benzyl 18 2-0Me-phenyl COH H H CH 19 phenyl CHzCHzCHs 20 phenyl N .~ _ _ OKi value followed by ASEM in parentheses. sent duplicate or triplicate determinations.

no. 7 8 9 10

u Ki, nMa 11440 (380) 11360 6150 1340 1150 11200 5850 2720

550

>loo00 10480 14280 1980 74 Ki values repre-

We have recently demonstrated that W(3-phenylpropyl)-l-phenyl-2-aminopropane(PPAP; 6, R = (6) Itzhak, Y. Pharmacological specificity of some psychotomimetic and antipsychotic agents for the sigma and PCP binding sites. Life Sci. 1988, 42, 745-752. (7) Itzhak, Y.; Kassim, C. 0. Clorgyline displays high affinity for u binding sites in C57BL/6 mouse brain. Eur. J. Pharmacol. 1990,176, 107-108. (8) Schmidt, A.; Lebel, L.; Koe, B. K.; Seeger, T.; Heym, J. Sertraline potently displaces (+)-["]3-PPP binding to u sites in rat brain. Eur. J. Pharmacol. 1989,165, 335-336. (9) Contreras, P. C.; Bremer, M. E.; Gray, N. M. Ifenprodil and SL 82.0715 potently inhibit binding of [3HH](+)-3-PPPto u binding sites in rat brain. Neurosci. Lett. 1990,116,190-193. (10) Lobner, D.; Lipton, P. u-Ligands and non-competitive NMDA antagonists inhibit glutamate release during cerebral ischemia. Neurosci. Lett. 1990, 117, 169-174. (11) Paul, I. A.; Kuypers, G.; Youdim, M.; Skolnick, P. Polyamines non-competitvely inhibit [3H]3-PPPbinding to sigma receptors. Eur. J. Pharmacol. 1990,184,203-204. (12) Iyengar, S.; Dilworth, V. K.; Mick, S. J.; Contreras, P. C.; Monahan, J. B.; Rao, T. S.; Wood, P. L. Sigma receptors modulate both A9 and A10 dopaminergic neurons in the rat brain: functional interaction with NMDA receptors. Brain Res. 1990,524, 322-326. (13) Junien, J.-L.; Gue, M.; Pascaud, X.; Fioramonti, J.; Bueno, L. Selective stimulation of colonic motor response to a meal by u ligands in dogs. Gastroenterology 1990,99,684-689. (14) Riviere, P. J. M.; Pascaud, X.; Junien, J. L.; Porreca, F. Neuropeptide Y and JO 1784, a selective u ligand, alter intestinal ion transport through a common, haloperidol-sensitive site. Eur. J. Pharmacol. 1990,187, 557-559. (15) Abreu, P.; Sugden, D. Characterization of binding sites for [3H]-DTG,a selective sigma receptor ligand, in the sheep pineal gland. Biochem. Biophys. Res. Commun. 1990, 171, 875-881. (16) Matsumoto, R. R.; Walker, J. M. Inhibition of rubral neurons by a specific ligand for u receptors. Eur. J.Pharmacol. 1988, 158, 161-165. (17) Goldstein, S. R.; Matsumoto, R. R.; Patrick, R. L.; Bowen, W. D.; Walker, J. M. Motor effects of two sigma ligands mediated by nigrwtriatal dopamine neurons. Synapse 1989,4,254-258. (18) Holtzman, S. J. Opioid- and phencyclidine-likediscriminative stimulus effects of ditolylguanidine, a selective sigma ligand. J.Pharmacol. Exp. Ther. 1989,248,1054-1062. (19) Wikstrom, H.; Andersson, B.; Elebring, T.; Svensson, K.; Carlsson, A.; Largent, B. N.-Substituted 1,2,3,4,4a,5,6,10boctahydrobenzo[f'Jquinolinesand 3-phenylpiperidines: Effects on central dopamine and u receptors. J. Med. Chem. 1987,30, 2169-2174.

NANM) and selectivity,21v22and have proposed that amine-substituted 2-phenylaminoethane constitutes the primary pharmacophore of the u opiates.22 This moiety is also present in the 3-phenylpiperidines, such as 3-PPP (31, and may constitute a common pharmacophore. In the course of our investigations on other neurotransmitter systems, we have previously shown that 4-phenylpiperazines occasionally serve as biological mimics of the phenylaminoethane moiety;23therefore, the possibility exista that appropriately substituted 1-phenylpiperazines might mimic u-related phenylaminoethanes and also bind a t u sites in a manner similar to that of 6. Some encouragement for this concept was derived from subsequent reports that piperazine derivatives such as azaperone (5)3 and BMY 14802 (a-(4-fluorophenyl)-4-(5-fluoro-2-pyrimidiny1)-1-piperazinebutanol)" bind with modest affinity at u receptors. Our investigation began with an evaluation of several simple 1-phenylpiperazine derivatives in order to (i) obtain their u affinity (Kivalues), (ii) formulate structure-affinity relationships, and (iii) determine how various substituenta influence binding. Many of these compounds (i.e., those shown in Table I) were already on hand from previous studies and offered a convenient starting point. Once these studies were accomplished, terminal amine substituents we had previously shown to be important for binding in the 6 series21were incorporated into the simple phenylpiperazine derivatives.

Chemistry Most of the target compounds were readily prepared in two or three steps (Table 11). For the most part, the syntheses involved acylation of a phenylpiperazine or phenylpiperidine with the appropriate acyl halide followed by reduction of the resulting amide with a BH3-Me2S mixture (method A), LiAlH4 (method B), or AH3 (method C). In several instances, a phenylpiperidine was reductively alkylated by treatment with 3-phenylpropionaldehyde under conditions of catalytic hydrogenation (method D). Other methods of preparation included formation of a piperidine ring by aminoalkylation of an a-methylstyrene using a method developed by Schmidle and M a n ~ f i e l d ~(method ~ l ~ ~ E), and acylation of a pre~~

(20) de Costa, B. R.; Rice, K. C.; Bowen, W. D.; Thurkauf, A.; Rothman, R. B.; Band, L.; Jacobson, A. E.; Radesca, L.; Contreras, P. C.; Gray, N. M.; Daly, I.; Iyengar, S.; Finn, D. T.; Varizani, S.; Walker, J. M. Synthesis and evaluation of Nsubstituted cis-N-methyl-2-(1-pyrrolidiny1)cyclohexylaminesas high affinity u receptor ligands. Identification of a new class of highly potent and selective u reeptor probes. J.Med. Chem. 1990,33, 3100-3110. (21) Glennon, R. A.; Smith, J. D.; Ismaiel, A. M.; El-Aahmawy, M.; Battaglia, G.; Fischer, J. B. Identification and exploitation of the u-opiate pharmacophore. J. Med. Chem. 1991, 34, 1094-1098. (22) Glennon, R. A.; Ismaiel, A. M.; Smith, J. D.; Yousif, M.; ElAshmawy, M.; Hemdon, J. L.; Fischer, J. B.; Burke Howie, K. J.; Binding of substituted and conformationallyrestricted derivatives of N-(3-phenyl-n-propyl)-l-phenyl-2-aminopropane (PPAF') at u sites. J.Med. Chem. 1991,34, 1855-1859. (23) Glennon, R. A.; Westkaemper, R. B.; Bartyzel, P. Medicinal chemistry of serotonergic agents In Serotonin Receptor Subtypes, Peroutka, s. J., Ed.; Wiley-Lisa: New York, 1991; pp 19-64. (24) Taylor, D. P.; Dekleva, J. Potential antipsychotic BMY 14802 selectively binds to sigma sites. Drug Deu. Res. 1987, 11, 65-70.

3362 Journal of Medicinal Chemistry, 1991, Vol. 34, No. 12

Table 11. Physicochemical and

u

Binding Properties of Piperidine and Piperazine Analogues k

no.

X N N N N N N N CH CH CH COH C= N N

-0(CH,)n-Ph

formulab u Ki,nMc ClsH21N2*HCld 10.5 (2.6) C1gH21F3N2*HCle 7.5 (1.4) 3 C1gH24N2*HClf 6.7 (0.7) 3 C1gH&1N2*HCld 19.2 (1.0) 3 C1gH23ClNz*HCP 2.9 (0.2) 4 C ~ O H & ~ * ~ H C ~ ~2.6 * ~(0.2) 4 C21H25F3NZ.HClJ 5.4 (1.2) 3 1.1 (0.2) 4 C21H2,N-HCl' 0.8 (0.1) 5 C22H2gN*HClk 0.9 (0.1) 3 CzoH25NO.HCl 3.0 (0.3) 4 C25HzsClNO: 4.1 (2.0) 3 C23H26Nz.HC1 2.8 (0.1) 4 CZiH26NzO.HCl 125 (17) 4.0 (1.4) (haloperidol) 10.4 (3.8) aPreparation = prep. First column is method (see Experimental Section), second is % yield, and third is recrystallization solvent: A = absolute EtOH, B = 2-butanone, E = absolute EtOH/anhydrous EhO, P = 2-propanol. bEmpiricalformula; all compounds analyzed within 0.4% of theory for C, H, and N. Compound 26 has been previously reported, but due to confusion about the salt (i.e., mono- or dihydrochloride), it was submitted for C1 analysis. 'Ki value is followed by hSEM. Each Ki value represents two to nine determinations. dCrystallized with 1mol of HzO. 'Crystallized with 0.5 mol of HzO. 'Crystallized with 1.25 mol of H20. g Compound 25, though mentioned in the literature (ref 29), has not been previously characterized. hCrystallized with 5 mol of H20;lit.30mp 205 "C. mp 214-216 OC. jcrystallized with 0.2 mol of H20. kCrystallized with 0.1 mol of HzO. 'Compound 32 represents a hydrogen maleate salt. 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 4

Ar 3-C1-phenyl 3-CF3-phenyl phenyl 3-C1-phenyl 4-C1-phenyl phenyl 3-CF3-phenyl phenyl phenyl phenyl phenyl 4-C1-phenyl 2-naphthyl benzoyl

Glennon et al.

n 2 2

A A B A C B B D B B D E B F

prepn 40 30 50 67 40 45 62 58 28 46 90 35 61 50

formed 1-(arylalky1)piperazine (method F).

Results and Discussion To initiate this investigation, we first examined the binding affinities of several simple 1-phenylpiperazines. 1-Phenylpiperazine (7) itself binds at u receptors with low affmity (Ki = 11440 nM; Table I) and with approximately 5 times the affiity of the parent 1-phenyl-2-aminopropane (6, R = H; Ki = 46400 nM).21 Realizing that l-phenyl-2aminopropane is a primary amine, that 1-phenylpiperazine is a secondary amine, and that simple N-methylation of 1-phenyl-2-aminopropane to the secondary amine 6 (R = CH3;Ki = 8320 nM) enhances affinity by about there is good agreement between the affmities of the parent structures for the two series of compounds. 1-Benzoylpiperazine (8) binds with an affinity (Ki = 11360 nM; Table I) comparable to that of 1-phenylpiperazine (7). Next, we examined a small series of 1-phenylpiperazine derivatives (already on hand as a result of other investigations) in order to determine whether or not aromatic substitution would have any significant affect on affinity. It is fairly evident from the results presented in Table I that most aromatic substituents have little effect. With the exception of the incorporation of a 4-ChlOrO group (Le., 11; Ki = 1150 nM) or a 3-trifluoromethyl group (Le., 10; Ki = 1340 nM), none of the other substituted simple piperazines (i.e., 9-13) displayed more than twice the affmity of the parent compound. Replacement of the phenyl portion of 7 with a 2-naphthyl group resulted in a 20-fold enhanced affinity (15; Ki = 550 nM), whereas replacement by 1-naphthyl (Le., 14) had less of an effect. Replacement of the piperazine 1-position nitrogen atom with a C(0H) group appears to have no effect on affiity (i.e., comparing 12 with 18), whereas replacement by a methylene group (comparing 4-phenylpiperidine (19; Ki = 1980 nM) with (25) Schmidle, C. J.; Mansfield, R. C. The aminomethylation of olefins. 11. A new synthesis of l-alkyl-4aryl-4-piperidinols. J. Am. Chem. SOC.1955, 77,5698-5700. (26) Schmidle, C. J.; Mansfield, R. C. The aminomethylation of olefins. 111. The synthesis of l-methyl-4-phenyl-4-acetoxypiperidine. J. Am. Chem. SOC.1955, 77, 5754-5755.

B B P B E E E E E E E E E E

mp, "C 197-198 200dec 178-180 168-169 196-197 208-210 148-149 217-218' 223-225 188-189 207-209 162-164 213-214 217-218

1-phenylpiperazine (7) resulted in about a 6-fold increase in affinity. N-Alkylation of the 1-phenyl-2-aminopropanes results in dramatic enhancement of u affinity.21s22Likewise, Nalkylation enhances the affmity of 1-phenylpiperazine. For example, incorporation of an n-propyl group, as in 1phenyl-4-n-propylpiperazine (20; Ki = 74 nM), results in increased aff~ty.Paralleling the approximately 1700-fold increase observed upon going from 1-phenyl-2-aminopropane (6, R = H) to PPAP (6, R = CH2CH2CH2CBH6), incorporation of an N-(3-phenylpropyl) group (Le., 23; Ki = 6.7 nM) also enhances the u affinity of l-phenylpiperazine by 1700-fold. On the basis of the data shown in Table 11, the 3- and 4-chloro derivatives of 23 were prepared and examined; the 3-chloro derivative 24 binds with approximately one-third the affinity of the parent compound 23, whereas the 4-chloro derivative 25 binds with about twice the affinity of 23. Shortening the side chain from three to two methylene units (Le., 21 and 22) results in little change in affinity, as does lengthening the chain from three to four methylene units (i.e., 26 and 27). The finding that the piperidine derivative 19 (Table I) binds with a several-fold greater affinity than the corresponding 1-phenylpiperazine (7) prompted the examination of several additional compounds. The tetrahydropyridyl analogue 32 binds with high affmity (Ki = 4.1 nM). Replacement of the N1 piperazine nitrogen atom of 23 (Ki = 6.7 nM) with a C(0H) group (Le., 31; Ki = 3.0 nM) doubles affinity. However, replacement of that same nitrogen atom by a CH group (28; Ki = 1.1nM) results in a 6-fold increase in affinity. This increase parallels the increase observed when comparing 7 and 19. Extending the side chain of 28 from three to either four (29; Ki = 0.8 nM) or five (30; Ki = 0.9 nM) methylene units has relatively little additional effect. It should be noted, however, that as with the 1-phenyl-2-aminopropane and l-phenylpiperazine series, the affinity of the N-(3-phenylpropyl) analogue 28 is 1700-fold higher than that of the parent 4-phenylpiperidine (19). Because increased lipophilicity of the terminal amine substituent seems to enhance the affinity of l-phenylpiperazines, we evaluated the N- (2-naphthylpropyl) ana-

Novel 1 -Phenylpiperazine and 4-Phenylpiperidine Derivatives

Journal of Medicinal Chemistry, 1991, Vol. 34,No. 12 3363

logue of 23 (i.e., 35; Ki= 4.0 nM); however, this compound was found to bind with an affiiity similar to that of 23 (Ki = 6.7 nM). 1-Benzoylpiperazine (8) binds at u receptors with low affinity (Ki> 11360 nM). Nevertheless, because 1-phenylpiperazine also binds with low affinity, we prepared the N-(Cphenylbutyl) derivative of 8 (34; Ki= 125 nM), which binds with only 100 times the affinity of the parent piperazine 8 and about one-fiftieth the affinity of 26. It would appear that incorporation of the N-(phenylalkyl) group has different effects on the affinity of 1benzoylpiperazine relative to 1-phenylpiperazine and 4phenylpiperidine, suggesting that the former may bind in a manner that differs somewhat from that of the latter two. While our studies were in progress, Largent and coworkers suggested that 3- and 4-phenylpiperidines constitute important pharmacophores at Q binding sites.3 Although considerable work has been done on 3-phenylpiperidine derivative^,^^^^' relatively little has been published on 4-phenylpiperidines. The results of the present study would tend to support the concept that 4-phenylpiperidines are important pharmacophores. Nevertheless, early reports led to molecular modeling studies involving certain phenylpiperidines and related compounds28that have resulted in hypotheses concerning a requirement for (or, at least, the importance of) electronegative substituents in the aromatic ring. For the most part, however, analogues lacking these electronegative substituents were not included in those studies. The results of the present investigation (as well as results of studies on analogues of 6)22 would suggest that although these electronegative substituents may be involved in binding, their role is minimal. For example, introduction of a 3-chloro group or a 4-chloro group enhances the affinity of l-phenylpiperazine (7) by 2-fold and 10-fold, respectively. However, for the N-(3-phenylpropyl) derivatives, incorporation of a 3-ChlOrO group (i.e., 24) reduces a f f i ~ t yby 3-fold, whereas incorporation of the 4-chloro group (i.e., 25) increases affiiity by only 2-fold. These aromatic substituents seem to have a less pronounced effect on the N-substituted derivatives than on their N-unsubstituted parent. Likewise, replacing the phenyl group of 1-phenylpiperazine (7) with a 2-naphthyl group increases affinity by eO-fold, whereas the naphthyl analogue 33 possesses only twice the affinity of its phenyl counterpart 23. The above results indicate that either (a) the simple phenylpiperazines are binding in a manner that is different from that of their N-substituted analogues, (b) the phenylpiperazines are behaving in a manner different from those compounds previously examined, or (c) the N substituent has a much more pronounced effect that overshadows the smaller effect of the aromatic substituents.

Table 111. Binding Profiles for Selected Examples

Van de Waterbeemd, H.; El Tayar, N.; Testa, B.; Wikstrom, H.; Largent, B. Quantitative structure-activity relationships and eudismic analyses of the preaynaptic dopaminergic activity and dopamine D 2 and u receptor affinities of 3-(3-hydroxypheny1)piperidines and octahydrobenzo[flquinolines.J.Med. Chem. 1987,30, 2175-2181. Manallack, D.T.; Beart, P. M. Quantitative conformational analyses predict distinct receptor sites for PCP-like and u drugs. Eur. J. Pharmacol. 1987,144, 231-235. Van den Brink, F.; Lien, E. J. pD2-,pA2-and pD,,-valuesof a series of compounds in a histaminic and a cholinergic system. Eur. J. Pharmacol. 1977,44, 251-270. Morren, H.; Zivkovic, D.;Linz, R.; Strubbe, H.; Marchal, L. Novel derivatives of N,"-disubstituted piperazinesand their neurotropic properties. Ind. Chim. Belge 1963, 2, 123-134. Arundel, J. H.;Somers, T. C.; Handley, G. J. Activity of a series of piperidines against Haemonchus contortus larvae. J. Med. Chem. 1966,9, 555-558.

ICw, nMa no. D1 D2 PCP 26 2820 (1450) 45 (*lo) >loo00 335 (*40) 28 >loo00 >loo00 29 >loo00 165 (*40) >loo00 30 >loo00 210 (*45) & m (o ;; 31 >loo00 1250 (*60) 35 3000 (*40) 160 (k30) >loo00 4 (halo) 337 (h95) 5.5 (h1.5) ND 'ICsovalues are followed by MEM. Values represent duplicate or triplicate determinationsexcept for haloperidol where n = 3-5. ND = not determined. Using radioligand binding assay techniques identical to those reported in the previous article in this series,22 compounds 31 and 35 were additionally examined at ,&adrenergic, muscarinic, quasqualate, and kainate binding sites and were found, in duplicate determinations, to bind with low affinity (ICso > loo00 nM).

One of the highest affinity u ligands and one involved in many of the molecular modeling studies is the piperidine derivative haloperidol (4). Although it has been assumed that certain structural features found in this molecule may be consequential to binding, the actual role of these substituents has not been specifically examined. It was not our intention to examine the structureaffinity relationships of haloperidol analogues; however, the results of the present study suggest that the previously published models may be in need of revision. For example, compound 29 may be viewed as an unadorned analogue of haloperidol that lacks (a) the aromatic fluoro group, (b) the carbonyl oxygen, (c) the benzylic hydroxyl group, and (d) the aro= 0.8 nM) binds with matic chloro group; and yet, 29 (Ki more than 10 times the affinity of haloperidol (Ki = 10.4 nM). Indeed, 29 is one of the highest affinity u ligands reported to date. Due to the structural similarity between 29 and haloperidol (41, it must be concluded that certain of the substituents in haloperidol actually detract from binding. A full understanding of the contribution of the other substituents awaits a detailed structureaffinity relationship investigation. Nevertheless, due to the higher affinity of 29 relative to haloperidol (4), it would appear that an N-substituted 4-phenylpiperidine is not only important for binding at u receptors3 but that it actually constitutes the primary u pharmacophore of haloperidol (present investigation). Furthermore, because 28 and 30 also bind with 10 times the affinity of 4, the four-carbon chain separating the phenyl group from the piperidine ring is not an absolute requirement for binding. Future molecular modeling studies will need to take these factors into account. Because various u ligands bind at dopamine D1 and D2 receptors and at PCP binding sites (see introduction), six representative high-affinity compounds were selected for further evaluation (Table 111). None of the compounds bind at PCP sites and all of the compounds bind at D1 and D2 sites with affinities significantly lower than that of haloperidol (4). Comparing the piperazine derivative 26 with its corresponding piperidine 29, it is clear that the piperidine analogue binds with a lower affinity at D1 and D2 dopamine receptors. Compounds 28-30 are piperidine derivatives that vary only with regard to their chain lengths; apparently, alteration of this chain length from three to five methylene groups has little effect on binding at dopamine receptors or PCP sites (Table 111). Nevertheless, these compounds bind at u receptors with greater than a 150-fold (D2 dopamine) to 10000-fold (D1 dopamine and PCP) selectivity. The benzylic hydroxy analogue 31 binds with a profile similar to that of its corresponding piperidine derivative 28 except that it possesses

3364 Journal of Medicinal Chemistry, 1991, Vol. 34, No. 12

a 4-fold lower affinity for D2 dopamine receptors. IC50 values for haloperidol binding at D1 and D2 receptors are provided for purpose of comparison (Table 111). Summary

Previous investigators have demonstrated that piperazine and piperidine derivatives bind at u receptors and have suggested that 4-phenylpiperidines may constitute a n important structural moiety for interaction at these sites. We have shown in the present study that 1phenylpiperazine, 4-phenylpiperidine, a n d simple aromatic-substituted derivatives thereof bind at u receptors with rather low affinity (Table I), but that modification of the terminal amine substituent so as to mimic the terminal amine Substituents of 6 can enhance affinity by more than 10000-fold. The 4-phenylpiperidine derivatives bind at u receptors with about 6 times the affinity of the corresponding 1-phenylpiperazines, a n d benzylic hydroxylation, although tolerated, does not enhance u affinity. Finally, removal of all the electronegative substituents of haloperidol results in a 10-fold increase in affinity; these findings provide information on a possible pharmacophore for haloperidol-type compounds a n d question the significance of previously published u binding models t h a t suggest that the presence of such electronegative substituents is important for binding. Experimental Section Chemistry. Proton magnetic resonance spectra were obtained with a JEOL FX9OQ spectrometer with tetramethylsilane as an internal standard. Spectral data are consistent with assigned structures. Melting points were determined with a ThomasHoover melting point apparatus and are uncorrected. Elemental analyses were performed by Atlantic Microlab and values are within 0.4% of theory. Experimental details below illustrate one of the methods (A-F) described in Table 11. 1-[3- ( T r i f l u o r o m e t h y l ) phenyl]-4- (2-phenyleth y l ) piperazine Hydrochloride (22). Method A. Phenylacetyl chloride (0.7 g, 4.3 mmol) in a 8 mL of dry THF was added in a dropwise manner to a stirred solution of freshly distilled 1[3-(trifluoromethyl)phenyl]piperazine(1.0 g, 4.3 "01) and EbN (0.4 g, 4.3 mmol) in dry THF at 0 "C. The reaction mixture was allowed to warm to room temperature with continued stirring (2 h). The solid was removed by filtration, and the filtrate was evaporated under reduced pressure to afford a clear oil. An EhO (100 mL) solution of the oil was washed with 5% HOAc (2 x 100 mL) and dried (MgSO,), and the solvents were removed at reduced pressure to afford a clear oil that solidified on standing to give 1.1g (73%) of the amide: mp 75-77 "C. A solution of the amide in dry THF (25 mL) was added in a dropwise (0.57 g, 1.6 "01) manner to a stirred solution of BH3-Me2S complex (15 mL of a 2.0 M solution in THF; 30 "01) in 20 mL of THF. The reaction mixture was allowed to stir at room temperature under an Nz atmosphere for 48 h and was then quenched by the addition of 25 mL of a saturated solution of HCl gas in MeOH and evaporated to a white residue. The residue was dissolved in MeOH (50 mL) and again evaporated to dryness. Concentrated HCl(l0 mL) was added, and the solvent was evaporated to yield a white residue. Recrystallization from 2-butanone afforded 0.18 g (30%) of 22: mp 200 "C dec. l-Phenyl-4-(3-phenylpropyl)piperazineHydrochloride (23). Method B. Hydrocinnamoyl chloride (1.5 mL, 0.01 mol) in THF (10 mL) was added dropwise to a stirred solution of 1-phenylpiperazine (1.5 mL, 0.01 mol) and Et3N (1.4 mL, 0.01 mol) in THF (40 mL). The reaction mixture was allowed to stir for 2 h and set aside overnight. The mixture was filtered and the solid material was washed with THF (ca. 30 mL). The combined filtrate and washings were evaporated under reduced pressure, and the oily residue was dissolved in Et20 (40 mL), washed with H 2 0 (2 X 20 mL), and dried (MgS04). The solvent was removed under reduced pressure to give a pale yellow oil of the amide intermediate in 70% yield. A solution of the amide (2.0 g, 7 mmol) in anhydrous EgO (30 mL) was added dropwise at 0 "C to a stirred suspension of LiAlH, (1.0 g, 25 mmol) in

Glennon et al. anhydrous EhO (30 mL). After complete addition, the reaction mixture was heated at reflux with stirring overnight and cooled to 0 "C, and excess LiAlH, was decomposed by addition of few drops of 30% NaOH. The reaction mixture was filtered and the solid material was washed with EhO (ca 30 mL). The combined filtrate and washings were washed with H20 (30 mL), dried (MgS04),and saturated with HCl gas to give a white solid hydrochloride salt; recrystallization from 2-propanol gave a 50% yield of compound 23: mp 178-180 "C. 1-( 4 4 hlorop henyl)-4-(3-phenylpropyl) piperazine Hydrochloride (25). Method C. Hydrocinnamoyl chloride (1.0 mL, 6.5 mmol) in THF (10 mL) was added dropwise to a stirred solution of 1-(4-~hlorophenyl)piperazine dihydrochloride (1.7 g, 6.5 mmol) and NEt3 (2.8 mL, 20 mmol) in THF (40 mL). The reaction mixture was allowed to stir overnight and filtered, and the solid material was washed with THF (ca. 30 mL). The combined filtrate and washings were evaporated under reduced pressure, and the oily residue was dissolved in EtzO (40 mL), washed with HzO (2 X 20 mL), and dried (MgSO,). The solvent was removed under reduced pressure to give a solid product (mp 112-113 "C) of the amide intermediate. A solution of AlH3 in anhydrous EhO was prepared by addition of LiA1H4 (2.5 g, 66 "01) to a stirred solution of AlC13 (2.9 g, 22 "01) in anhydrous EhO (50 mL). Immediately thereafter, a solution of the above in a mixture of anhydrous EbO (25 mL) amide (0.73 g, 22 "01) and dry THF (5 mL) was added dropwise with stirring. The reaction mixture was allowed to stir Overnight at room temperature and then heated gently at reflux for 3 h. The reaction flask was cooled in an ice bath, and excess hydride was decomposed by addition of wet THF (5 mL) and 2% NaOH solution (5 mL). The residual oil was dissolved in E g o (50 mL), washed with HzO (2 X 30 mL), dried (MgSO,), and saturated with HCl gas. The hydrochloride salt was collected by filtration, dried, and recrystallized from an absolute EtOH-EhO mixture to afford 0.28 g (40%) of 25 as white crystals: mp 196-197 "C. 4-Hydroxy-4-phenyl-l-(3-phenylpropyl)piperidineHydrochloride (31). Method D. A mixture of 4-hydroxy-4phenylpiperidine (0.35 g, 2 "01) and hydrocinnamaldehyde (0.27 g, 2 mmol) in 95% EtOH (20 mL) was hydrogenated at 45 psi in the presence of 10% Pd/C (100 mg) for 2 h. The suspension was filtered, the filtrate was evaporated under reduced pressure to an oily residue, and an ethereal solution of the oil was treated with HCl gas-saturated EhO until precipitation ceased. The crude product was collected by filtration and recrystallized from an absolute EtOH-anhydrous EhO mixture to yield 0.59 g (90%) of 31 as white crystals: mp 207-209 "C. 4-(4-Chlorophenyl)-l-(4-phenylbutyl)-l~,5,6-tetrahydropyridine Hydrogen Maleate (32). Method E. Orthophosphoric acid (85% 0.43 g) was slowly added to solution of glacial HOAc (1.9 g) and AczO (0.4 9); after the exothermic reaction ceased, paraformaldehyde (0.36 g), 4-chloro-a-methylstyrene (0.62 g, 4 mmol), and Cphenylbutylamiie (0.6 g, 4 "01) were added. The reaction mixture was allowed to stir at 100-115 "C for 4 h and was then set aside for 2 days. Water (10 mL) was added; the mixture was made basic by the addition of solid Na2C03and extracted with hexanes (3 X 10 mL). The solution was dried (anhydrous KZCOJ, and the solvent was evaporated. The crude residue was recrystallized from MeOH to give about 0.4 g of the free base (mp 87-90 "C). A solution of the free base in anhydrous EhO was treated with a saturated ethereal solution of maleic acid (0.5 g/20 mL), and the crude salt was collected and recrystallized from absolute EtOH-anhydrous EgO to give 0.6 g (35%) of 32 as fine crystals: mp 162-164 OC. l-Benzoyl-4-(4-phenylbutyl)piperazine Hydrochloride (34). Method F. A solution of N-(4phenylbutyl)piperazine (36) (1.0 g, 4.6 mmol) in CHC1, (20 mL) was added to a solution of benzoyl chloride (0.7 g, 5 mmol) in CHC13 (20 mL) at 0 "C. The reaction mixture was allowed to stir at room temperature overnight, the solvent was removed by evaporation under reduced pressure, and the yellow solid residue was recrystallized from an absolute EtOH-anhydrous EgO mixture to afford 0.83 g (50%) of 34: mp 217-218 "C. l-Pheny1-4-[3-(2-naphthyl)propyl]piperazineDihydrochloride (35). Ethyl chloroformate (0.35 g, 3 mmol) was added dropwise at 0 "C to stirred solution of 3-(2-naphthyl)propionic acid (0.6 g, 3 mmol) and NE5 (0.3 g, 3 "01) in CH2C12(30 mL).

Novel 1-Phenylpiperazine and 4-Phenylpiperidine Derivatives

Journal of Medicinal Chemistry, 1991, Vol. 34, No. 12 3365

The reaction mixture was allowed to stir at 0 "C for 30 min, whereupon a solution of 1-phenylpiperazine (0.5 g, 3 mmol) in CHzClz(10 mL) was added in dropwise manner. After complete addition, the reaction mixture was allowed to stir at room temperature for 4 h, washed with water (2 X 10 mL), and dried (MgSO,). The solvent was removed by evaporation under reduced pressure to give a white solid residue; recrystallization from 95% EtOH gave 0.7 g (66%) of the amide as white crystals (mp 101 "C). A solution of the amide (0.68 g, 2 mmol) in dry THF (30 mL) was added dropwise at 0 "C to stirred suspension of LiAlH, (0.5 g, 12 mmol) in dry THF (30 mL). After complete addition, the stirred reaction mixture was heated at reflux overnight. The mixture was cooled to 0 "C and the excess L W 4 was decomposed by addition of HzO containing a few drops of 30% NaOH. The reaction mixture was filtered and the solid material was washed with THF (ca 20 mL). The combined filtrate and washings were evaporated under reduced pressure; an EhO (40 mL) solution of the oily residue was washed with HzO (20 mL), dried (MgSO,), and saturated with HCl gas to give a white solid hydrochloride salt. Recrystallization from absolute EtOH gave 0.48 g (66%) of 35 as white crystals: mp 190-192 "C. Anal. (Cz3HzsNz.2HC1)

inorganic material was removed by filtration and washed with THF (ca. 30 mL). The combined filtrates and washings were evaporated in vacuo to dryness. The residual oil was dissolved in EhO (40 mL), washed with HzO (20 mL), and dried (MgS04). After evaporation of the solvent under reduced pressure, 1.4 g (71%) of a clear oil corresponding to 1-(4-phenylbutyl)piperazine was obtained. (A sample of the HC1 salt was prepared for identification: mp 242-245 "C after recrystallization from EtOH-EhO.) The free base was used without further purification in the preparation of 34. Radioligand Binding Studies. Guinea Pig/ [3H]DTG. Sigma receptor binding assays, using [3H]DTG (di-o-tolylguanidine) as radioligand and guinea pig brain membranes as source of receptor, were performed exactly as described in our previous study.22 Briefly, guinea pig brain membranes (P2 microsomal fraction) were prepared from frozen guinea pig brains (Taconic) to a f i d protein concentration of 3 mg/mL and stored at -70 "C. For the assay, the membranes were thawed and diluted 1:3 with 50 mM Tris-HC1 (pH 7.41, and 0.4 mL was combined with 50 pL of [3H]DTG(1-2 nM final concentration) and 50 pL of unlabeled competing drug or buffer. The mixtures were incubated for 90 min at room temperature and incubation was terminated by rapid filtration under vacuum through Whatman GF/B or Schleicher & Schuell no. 32 glass fiber filters using a Brandel &well cell harvester. The fiiters were washed threetimes with 5 mL of cold Tris-HC1 buffer, and each fiiter was suspended in 5 mL of Cytoscint (ICN Biomedicals); radioactivity was measured by liquid scintillation spectrometry at a counting efficiency of 50%. Nonspecific binding was measured in the presence of 10 gM haloperidoL Data represent the mean and SEM of at least three competition curves (unleea otherwise stated). IC50 values, determined by analyzing displacement curves using nonlinear least-squares regression analysis, were converted to Ki values using the Cheng-Prusoff equation.32 Other Assays. Dopamine D1 ([3H]SCH-23390),dopamine D2 ([3H]domperidone),and PCP ( [3H]MK-801)receptor assays were also performed as described in the previous manuscript in this series.22 Dopamine binding assays used washed membranes prepared from frozen rat striata resuspended in a buffer containing 50 mM Tris-HCl,120 mM NaCl, 5 mM KC1,2 mM CaClZ,and 1 mM MgClz (pH 7.4 at 37 "C). PCP binding assays were performed with rat brain membranes using [3H]MK-801(97 Ci/ mmol)

c1.

N-(4-Phenylbutyl)piperazine (36). Ethyl chloroformate (4.32 g, 0.04 mol) was added dropwise, at 0 "C, to a stirred solution of 4-phenylbutyric acid (4.56 g, 0.04 mol) and NEt3 (4.0 g, 0.04 mol) in CHzCl2(40 mL); stirring was allowed to continue for 30 min at 0-5 "C, whereupon a solution of l-piperazinecarboxaldehyde (4.56 g, 0.04 mol) in CHzClz(20 mL) was added. The reaction mixture was allowed to stir at room temperature for 4 h and set aside overnight. The mixture was washed with HzO (2 X 20 mL) and the organic portion was dried (MgS04) and evaporated in vacuo to drynesa. The residual oil (5.8 g, 80%)was used in the next step without further purification. The N-formyl derivative (5.7 g, 22 mmol) was dissolved in a concentrated HC1-methanol solution (1:9,200 mL). The resulting mixture was stirred at room temperature for 18 h, and the solvent was then removed by evaporation under reduced pressure. The solid residue was suspended in absolute EtOH (200 mL), heated under reflux for 15 min, and filtered while hot. After cooling to room temperature, the alcoholic solution was diluted with anhydrous EhO and allowed to stand at -15 "C overnight. The white precipitated product was collected by filtration and air-dried to give 2.3 g (40%) of 1-(4phenylbutyryl)piperazinehydrochloride (mp 185-188 "C). The free base of the amide (2 g, 8.6 mmol) was dissolved in dry THF (40 mL) and added in dropwise manner, at 0 "C, to a stirred suspension of L W 4 (2.0 g, 52 "01) in dry THF (40 mL). After complete addition, the reaction mixture was heated at reflux overnight and then cooled to 0 "C. Exceas L W 4was decomposed by addition of HzO (2 mL) and 30% NaOH (5 drops). The

.

(32) Cheng, Y. C.; Prusoff, W. H. Relationship between the inhibition constant (Ki) and the concentration of inhibitor which causes 50% inhibition (IC50) of an enzymatic reaction. Biochem. Pharmacol. 1973,22, 3099-3018.